A Method to Classify Steel Plate Faults Based on Ensemble Learning

Abstract

With the industrial revolution 4.0, machine learning methods are widely used in all aspects of manufacturing to perform quality prediction, fault diagnosis, or maintenance. In the steel industry, it is important to precisely detect faults/defects in order to produce high-quality steel plates. However, determining the exact first-principal model between process parameters and mechanical properties is a challenging process. In addition, steel plate defects are detected through manual, costly, and less productive offline inspection in the traditional manufacturing process of steel. Therefore, it is a great necessity to enable the automatic detection of steel plate faults. To this end, this study explores the capabilities of the following three machine learning models Adaboost, Bagging, and Random Forest in detecting steel plate faults. The well-known steel plate failure dataset provided by Communication Sciences Research Centre Semeion was used in this study. The aim of many studies using this dataset is to correctly classify defects in steel plates using traditional machine learning models, ignoring the applicability of the developed models to real-world problems. Manufacturing is a dynamic process with constant adjustments and improvements. For this reason, it is necessary to establish a learning process that determines the best model based on the arrival of new information. Contrary to previous studies on the steel plate failure dataset, this article presents a systematic modelling approach that includes the normalization step in the data preparation stage to reduce the effects of outliers, the feature selection step in the dimension reduction stage to develop a machine learning model with fewer inputs, and hyperparameter optimization step in the model development stage to increase the accuracy of the machine learning model. The performances of the developed machine learning models were compared according to statistical metrics in terms of precision, recall, sensitivity, and accuracy. The results revealed that AdaBoost performed well on this dataset, achieving accuracy scores of 93.15% and 91.90% for the training and test datasets, respectively.

Keywords

• : Machine learning
• Classification
• Ensemble methods
• Fault detection
• , Fault detection
• Artificial Learning

Toplu Öğrenmeye Dayalı Çelik Levha Arızalarını Sınıflandırması İçin Bİr Yöntem

Abstract

Endüstri devrimi 4.0 ile birlikte, makine öğrenimi yöntemlerini, kalite tahminini, arıza tespitini veya bakımını gerçekleştirmek için üretimin her alanında yaygın olarak kullanılmaktadır. Çelik endüstrisinde, yüksek kaliteli çelik levhalar üretmek için arızaları/kusurları tam olarak tespit etmek önemlidir. Ancak, süreç parametreleri ve mekanik özellikler arasındaki kesin birinci ana modeli belirlemek zorlu bir süreçtir. Ek olarak, çelik levha kusurları, çeliğin geleneksel üretim sürecinde manuel, maliyetli ve daha az üretkenlik sağlayan çevrimdışı denetim yoluyla tespit edilir. Bu nedenle, çelik plaka hatalarının otomatik olarak tespit edilmesini sağlamak büyük bir zorunluluktur. Bu amaçla, bu çalışma, aşağıdaki üç grup makine öğrenme modeli Adaboost, Bagging ve Random Forest'ın çelik levha hatalarının tespitinde yeteneklerini araştırmaktadır. Bu çalışmada İletişim Bilimleri Araştırma Merkezi Semeion tarafından sağlanan iyi bilinen çelik levha arıza veri seti kullanılmıştır. Bu veri setini kullanan birçok araştırmanın amacı, geliştirilen modellerin gerçek dünya problemlerine uygulanabilirliğini göz ardı ederek, geleneksel makine öğrenimi modellerini kullanarak çelik levhalardaki hataları doğru bir şekilde sınıflandırmaktır. Üretim dinamik bir süreçtir, sürekli ayarlamalar ve iyileştirmeler yapılır. Bu nedenle yeni bilgilere dayalı olarak en iyi modeli belirleyen yapay öğrenme süreçlerinin oluşturulması gerekmektedir. Çelik levha arıza veri seti ile ilgili önceki çalışmaların aksine, bu makale, aykırı değerlerin etkilerini azaltmak için veri hazırlama aşamasında normalleştirme adımını, kısa hesaplama süresi olan bir makine öğrenmesi modeli geliştirmek için boyut küçültme aşamasında özellik seçimi adımını ve makine öğrenmesi modelinin doğruluğunu artırmak için model geliştirme aşamasında hiperparametre optimizasyon adımını içeren sistematik bir modelleme yaklaşımı sunmaktadır. Geliştirilen makine öğrenmesi modellerinin performansları istatistiksel ölçütlere göre kesinlik, geri çağırma, duyarlılık ve doğruluk açısından karşılaştırılmıştır. Sonuçlar, eğitim ve test veri kümeleri için sırasıyla %93.15 ve %91.90 doğruluk puanlarına ulaşan AdaBoost'un bu veri setinde en iyi performansı gösterdiğini ortaya koymuştur.

Keywords

• Fault detection
• Machine learning
• Classification
• Ensemble methods
• Fault detection

References

1. Alkan B., Bullock, S., Assessing operational complexity of manufacturing systems based on algorithmic complexity of key performance indicator time-series. Journal of the Operational Research Society 72(10), 2241-2255, 2021.
2. Backman J., Kyllönen V., Helaakoski H., Methods and tools of improving steel manufacturing processes: Current state and future methods. IFAC-PapersOnLine 52(13), 1174-1179, 2019.
3. Bektas O., Jones J. A., Sankararaman S., Roychoudhury I., Goebel K., Reconstructing secondary test database from PHM08 challenge data set. Data in Brief 21, 2464-2469, 2018.
4. Bektas O., Jones J. A., Sankararaman S., Roychoudhury I., Goebel K., A neural network framework for similarity-based prognostics. MethodsX 6, 383-390, 2019.
5. Buscema M., Terzi S., Tastle W., A new meta-classifier. Annual Meeting of the North American Fuzzy Information Processing Society (NAFIPS), Toronto, ON, Canada, July 12-14, 2010, pp: 1-7.
6. Ceryan N., Ozkat E. C., Korkmaz Can N., Ceryan S., Machine learning models to estimate the elastic modulus of weathered magmatic rocks. Environmental Earth Sciences 80(12), 1-24, 2021.
7. Chen J., The Application of tree-based ML algorithm in steel plates faults identification. Journal of Applied and Physical Sciences 4(2), 47-54, 2018.
8. Fakhr M., Elsayad A. M., Steel plates faults diagnosis with data mining models. Journal of Computer Science 8(4), 506-514, 2012.

9. Gamal M., Donkol A., Shaban A., Costantino F., Di G., Patriarca R., Anomalies detection in smart manufacturing using machine learning and deep learning algorithms. In Proceedings of the International Conference on Industrial Engineering and Operations Management, Rome, Italy, August 2-5, 2021, pp: 1611-1622.
10. Gao Y., Gao L., Li X., Yan X., A semi-supervised convolutional neural network-based method for steel surface defect recognition. Robotics and Computer-Integrated Manufacturing 61, 101825, 2020.
11. Kaggle, A., Faulty Steel Plates, Research Center of Sciences of Communication, https://www.kaggle.com/datasets/uciml/faulty-steel-plates, (Retrieved August 8, 2022), 2017.
12. Kahveci S., Alkan B., Musab H, A., Ahmad B., Harrison R., An end-to-end big data analytics platform for IoT-enabled smart factories: A case study of battery module assembly system for electric vehicles. Journal of Manufacturing Systems 63, 214-223, 2022.
13. Kazemi M. A. A., Hajian S., Kiani N., Quality Control and Classification of Steel Plates Faults Using Data Mining. Applied Mathematics Information Sciences Letters 6(2), 59-67, 2018.
14. Kharal A., Explainable artificial intelligence based fault diagnosis and insight harvesting for steel plates manufacturing. arXiv preprint arXiv:2008.04448, 2020.
15. Kurt R., Industry 4.0 in terms of industrial relations and its impacts on labour life. Procedia computer science 158, 590-601, 2019.
16. Lennox B., Montague G., Marjanovic O., Detection of faults in Batch Processes: Application to an industrial fermentation and a steel making process. Water Science and Technology, 2000.
17. Liu Y., Gao H., Guo L., Qin A., Cai C., You Z., A data-flow oriented deep ensemble learning method for real-time surface defect inspection. IEEE Transactions on Instrumentation and Measurement 69(7), 4681-4691, 2019.
18. Markoulidakis I., Rallis I., Georgoulas I., Kopsiaftis G., Doulamis A., Doulamis N., Multiclass Confusion Matrix Reduction Method and Its Application on Net Promoter Score Classification Problem. Technologies 9(4), 81, 2021.
19. Nkonyana T., Sun Y., Twala B., Dogo E., Performance evaluation of data mining techniques in steel manufacturing industry. Procedia Manufacturing 35, 623-628, 2019.
20. Ozkat E. C., Franciosa P., Ceglarek D., Laser dimpling process parameters selection and optimization using surrogate-driven process capability space. Optics & Laser Technology 93, 149-164, 2017a.
21. Ozkat E. C., Franciosa P., Ceglarek D., Development of decoupled multi-physics simulation for laser lap welding considering part-to-part gap. Journal of Laser Applications 29(2), 022423, 2017b.
22. Özkat E. C., Makine Öğrenmesi Metodolojisi Kullanılarak Yüksek Hızlı Rulmanlarda Sağlık Göstergesinin Belirlenmesi. Avrupa Bilim ve Teknoloji Dergisi (22), 176-183, 2021.
23. Pham T. A., Tran V. Q., Developing random forest hybridization models for estimating the axial bearing capacity of pile. Plos one, 17(3), e0265747, 2022.
24. Simić D., Svirčević V., Simić S., An approach of steel plates fault diagnosis in multiple classes decision making. In International Conference on Hybrid Artificial Intelligence Systems, Salamanca, Spain, June 11-13, 2014, pp: 86-97.
25. Srivastava A. K., Comparison analysis of machine learning algorithms for steel plate fault detection. International Research Journal of Engineering and Technology 6(4), 1231-1234, 2019.
26. Tasar B., Comparison Analysis of Machine Learning Algorithms for Steel Plate Fault Detection. Düzce Üniversitesi Bilim ve Teknoloji Dergisi 10(3), 1578-1588, 2022.
27. Tian Y., Fu M., Wu F. Steel plates fault diagnosis on the basis of support vector machines. Neurocomputing 151, 296-303, 2015.
28. Widodo A., Yang B. S., Support vector machine in machine condition monitoring and fault diagnosis. Mechanical systems and signal processing 21(6), 2560-2574, 2007.
29. Xiong J., Pang Q., Cheng W., Wang N., Yong Z., Reservoir risk modelling using a hybrid approach based on the feature selection technique and ensemble methods. Geocarto International 37(11), 3312-3336, 2022.
30. Xu L. D., Xu E. L., Li L. Industry 4.0: state of the art and future trends. International journal of production research 56(8), 2941-2962, 2018.
31. Yang K., Yu Z., Chen C. P., Cao W., Wong H. S., You J., Han G., Progressive hybrid classifier ensemble for imbalanced data. IEEE Transactions on Systems, Man, and Cybernetics: Systems 52(4), 2464-2478, 2021.
32. Zhang X., Kano M., Tani M., Mori J., Ise J., Harada K., Prediction and causal analysis of defects in steel products: Handling nonnegative and highly overdispersed count data. Control Engineering Practice 95, 104258, 2020.
33. Zhao Z., Yang J., Lu W., Wang X., Application of local outlier factor method and back-propagation neural network for steel plates fault diagnosis. In The 27th Chinese Control and Decision Conference (2015 CCDC), Qingdao, China, 23-25 May, 2015, pp: 2416-2421.